Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Measuring Torque on the Impeller Shaft of a Stirred Tank

  1. Sep 14, 2018 #1
    Hi Everyone,

    I am working on a project which requires me to determine the torque on the impeller shaft of stirred tank. Currently as I see it, there are 2 viable options:

    1. To measure the torque in the shaft using a torque transducer such as this one. This is by far the most accurate means of measurement, however the technique is extremely expensive. A torque cell and a DAQ to go along with it would cost around $3000.
    2. To measure the torque by placing the tank on a freely rotating platform (such as a lazy Susan), attaching a force transducer to the lazy Susan, allowing the impeller to spin and measuring the tangential force necessary to keep the tank from rotating. This could be used to calculate the force simply using the formula τ= r × F. This is a good option as well, however it is known that energy is lost in the form of heat within the fluid of a stirred tank as well as in the bearings of the lazy Susan.
    I understand torque measurement is also possible using strain gauges in a Wheatstone bridge configuration, however I have consulted with a few of my professors and it seems that this method lacks the robustness and accuracy necessary for our application. I am also aware that I could calculate the torque using the equation P=2πτN (Power = 2π × torque × rotational speed), however this method lacks significant accuracy as well.

    I have a couple of questions.

    Will the heat loss to the fluid be significant? Is there a way to calculate this heat loss? Additionally, does this “lazy Susan” method have a name? I have been looking for some information on this method both online and in the Industrial Fluid Mixing handbook, but to no avail. If anyone knows of any articles on it or simply the methods name to put me on the right track I would greatly appreciate it.

    Also, is there another way to measure shaft torque that I am not thinking of?

  2. jcsd
  3. Sep 14, 2018 #2


    User Avatar
    Gold Member

    You may want to look at the torque sensor/arrangements used on engine dynamometers.

    I would think you could be creative with your agitator mount, making it semi-fixed, with an appropriate length lever arm to a strain gauge. You may have to play with it a little to get the right range load cell with the right length mount.
  4. Sep 14, 2018 #3


    User Avatar
    Science Advisor

    You can measure the torque in the stirrer shaft, between the tank and the ground or between the stirrer mount and the strirrer drive.
    The stirrer shaft is the most difficut because it is rotating. Torque on the stirrer motor mount is easier.

    But the easiest way might be to use a DC electric motor to drive the stirrer, vary the voltage to adjust the RPM, then monitor the motor current which is a very good proxy for motor torque.
  5. Sep 15, 2018 #4

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Why are you worried about heat ?

    Your "Lazy Susan" sounds a lot like a dynamometer and is a good approach.
  6. Sep 16, 2018 #5


    User Avatar
    Gold Member

    I see some problem with this approach.
    The torque from the spinnor to the walls of vessel translate into a torque onto the vessel only when the fluid is moving at the walls.

    At start up for example the shaft will have a high torque as the spinnor begins to accelerate the fluid but the vessel will not yet experience any torque to it ( as there is no fluid movement along the wall, and thus no shear stress )
    Maximum torque on the vessel will be when the fluid is rotating as a bulk unit, in which case the shaft will experience its least amount of torque as it is no longer accelerating any of the fluid, but just keeping the fluid's rotational velocity constant.
  7. Sep 16, 2018 #6

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Acceleration of the fluid indeed won't show up at the wall, but once it reaches steady state viscous friction wlll make torque show up there.

    He could measure torque at the motor - that would be better anyway. Mount it on something that's free to rotate a bit and attach a scale.

  8. Sep 16, 2018 #7


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member


    It might be possible to mount the stirrer motor in such a way that you can measure the torque there.
  9. Sep 17, 2018 #8
    The responses above answer your precise question - I agree that measuring the torque at the motor with a motor mount which allows for that is probably your best (budget) option. That motor mount may not be trivial, depending on your error budget and the magnitude of the forces involved - you'll definitely get an education in the course of implementing it. The Omega part may not seem so expensive after that.
    You don't give enough application particulars (speed, etc) to be sure, but if it's the viscosity of the 'stirred' material that you're after, you'd probably be better served by an independent viscometer- that could be cheaper and more precise.

    Heat loss: If you assume that all of 'shaft power' produced by the motor becomes heat in the fluid. you won't miss by much. That wouldn't be a bad way to sanity-check your torque measurement (given RPM), if you can control the heat loss in the vessel.
  10. Sep 17, 2018 #9


    User Avatar
    Science Advisor

    You have a closed mechanical loop that must transmit the torque. It includes the stirrer, the motor, the motor mounts, the platform and the pot on the platform being stirred. Measurement of torque anywhere in that loop will give you the answer.

    I am surprised no-one has recognised the simplicity of monitoring electric motor current. Motor current is directly proportional to torque. The economy of production requires the optimum utilisation of machine tools to maximise process throughput. Motor current is monitored on grinders to regulate the rate of cut, and to limit the rate of heat generation during polishing processes.

    What type of motor is fitted to the stirrer now?
  11. Sep 18, 2018 #10


    User Avatar

    Staff: Mentor

    Since the OP is worried about it, I'd guess this is a large and heat sensitive process. I did process utilities for a cocoa processing plant a few years ago and they included jacket cooling for some 100 kW blenders. You wanted the cocoa to be liquid, but not to cook it.
  12. Sep 23, 2018 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    True enough for series wound brushed motors but not for AC induction motors.

    He asked if heat "loss" would be significant
    which made me wonder what he was really asking.
    If it's a stirrer doesn't all the torque go into heating the fluid ? As opposed to a pump which raises its pressure or velocity ?
    Why would paddlewheel work be considered a loss?
    That's all i was asking.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted